deepbox 0.1.0 → 0.2.0

package/dist/{index-DuCxd-8d.d.ts → index-D9ztTlDr.d.ts}

@@ -1,6 +1,16 @@
-import { T as Tensor } from './Tensor-g8mUClel.js';
-import { C as CSRMatrix } from './CSRMatrix-KzNt6QpS.js';
+import { T as Tensor } from './Tensor-BORFp_zt.js';
+import { C as CSRMatrix } from './CSRMatrix-CtD23fRM.js';
 
+/**
+ * Input type accepted by 1D encoder methods (LabelEncoder, LabelBinarizer).
+ * Accepts a Tensor directly or a plain JavaScript array of strings, numbers, or booleans.
+ */
+type EncoderInput1D = Tensor | readonly (string | number | bigint | boolean)[];
+/**
+ * Input type accepted by 2D encoder methods (OneHotEncoder, OrdinalEncoder).
+ * Accepts a Tensor directly or a plain JavaScript array of arrays.
+ */
+type EncoderInput2D = Tensor | readonly (readonly (string | number | bigint)[])[];
 /**
  * Type representing a category value that can be a string, number, or bigint.
  * Used for categorical encoding operations.
@@ -33,7 +43,7 @@ type CategoriesOption = "auto" | ReadonlyArray<ReadonlyArray<Category>>;
  * const yDecoded = encoder.inverseTransform(yEncoded); // ['cat', 'dog', 'cat', 'bird']
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html | Scikit-learn LabelEncoder}
+ * @see {@link https://deepbox.dev/docs/preprocess-encoders | Deepbox Encoders}
  */
 declare class LabelEncoder {
     /** Indicates whether the encoder has been fitted to data */
@@ -50,7 +60,7 @@ declare class LabelEncoder {
      * @returns this - Returns self for method chaining
      * @throws {InvalidParameterError} If y is empty
      */
-    fit(y: Tensor): this;
+    fit(y: EncoderInput1D): this;
     /**
      * Transform labels to normalized encoding.
      * Each unique label is mapped to an integer in [0, n_classes-1].
@@ -60,7 +70,7 @@ declare class LabelEncoder {
      * @throws {NotFittedError} If encoder is not fitted
      * @throws {InvalidParameterError} If y contains labels not seen during fit
      */
-    transform(y: Tensor): Tensor;
+    transform(y: EncoderInput1D): Tensor;
     /**
      * Fit label encoder and return encoded labels in one step.
      * Convenience method equivalent to calling fit(y).transform(y).
@@ -68,17 +78,17 @@ declare class LabelEncoder {
      * @param y - Target labels (1D tensor)
      * @returns Encoded labels as integer tensor
      */
-    fitTransform(y: Tensor): Tensor;
+    fitTransform(y: EncoderInput1D): Tensor;
     /**
      * Transform integer labels back to original encoding.
      * Reverses the encoding performed by transform().
      *
-     * @param y - Encoded labels (1D integer tensor)
+     * @param y - Encoded labels (1D integer tensor or number array)
      * @returns Original labels (strings or numbers)
      * @throws {NotFittedError} If encoder is not fitted
      * @throws {InvalidParameterError} If y contains invalid indices
      */
-    inverseTransform(y: Tensor): Tensor;
+    inverseTransform(y: EncoderInput1D): Tensor;
 }
 /**
  * Encode categorical features as one-hot numeric array.
@@ -105,7 +115,7 @@ declare class LabelEncoder {
  * // Result: [[1,0,1,0,0], [0,1,0,1,0], [1,0,0,0,1]]
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html | Scikit-learn OneHotEncoder}
+ * @see {@link https://deepbox.dev/docs/preprocess-encoders | Deepbox Encoders}
  */
 declare class OneHotEncoder {
     /** Indicates whether the encoder has been fitted to data */
@@ -150,7 +160,7 @@ declare class OneHotEncoder {
      * @throws {ShapeError} If X is not a 2D tensor
      * @throws {InvalidParameterError} If X is empty
      */
-    fit(X: Tensor): this;
+    fit(X: EncoderInput2D): this;
     /**
      * Transform X using one-hot encoding.
      * Each categorical value is converted to a binary vector.
@@ -160,8 +170,8 @@ declare class OneHotEncoder {
      * @throws {NotFittedError} If encoder is not fitted
      * @throws {InvalidParameterError} If X contains unknown categories
      */
-    transform(X: Tensor): Tensor | CSRMatrix;
-    fitTransform(X: Tensor): Tensor | CSRMatrix;
+    transform(X: EncoderInput2D): Tensor | CSRMatrix;
+    fitTransform(X: EncoderInput2D): Tensor | CSRMatrix;
     inverseTransform(X: Tensor | CSRMatrix): Tensor;
 }
 /**
@@ -187,7 +197,7 @@ declare class OneHotEncoder {
  * // Result: [[1, 1], [0, 0], [2, 1]] (alphabetically sorted)
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html | Scikit-learn OrdinalEncoder}
+ * @see {@link https://deepbox.dev/docs/preprocess-encoders | Deepbox Encoders}
  */
 declare class OrdinalEncoder {
     /** Indicates whether the encoder has been fitted to data */
@@ -223,7 +233,7 @@ declare class OrdinalEncoder {
      * @returns this - Returns self for method chaining
      * @throws {InvalidParameterError} If X is empty
      */
-    fit(X: Tensor): this;
+    fit(X: EncoderInput2D): this;
     /**
      * Transform X using ordinal encoding.
      * Each category is mapped to its index in the sorted categories array.
@@ -233,7 +243,7 @@ declare class OrdinalEncoder {
      * @throws {NotFittedError} If encoder is not fitted
      * @throws {InvalidParameterError} If X contains unknown categories
      */
-    transform(X: Tensor): Tensor;
+    transform(X: EncoderInput2D): Tensor;
     /**
      * Fit encoder and transform X in one step.
      * Convenience method equivalent to calling fit(X).transform(X).
@@ -241,7 +251,7 @@ declare class OrdinalEncoder {
      * @param X - Training data (2D tensor)
      * @returns Encoded data
      */
-    fitTransform(X: Tensor): Tensor;
+    fitTransform(X: EncoderInput2D): Tensor;
     /**
      * Transform ordinal integers back to original categories.
      * Reverses the encoding performed by transform().
@@ -251,7 +261,7 @@ declare class OrdinalEncoder {
      * @throws {NotFittedError} If encoder is not fitted
      * @throws {InvalidParameterError} If X contains invalid indices
      */
-    inverseTransform(X: Tensor): Tensor;
+    inverseTransform(X: EncoderInput2D): Tensor;
 }
 /**
  * Binarize labels in a one-vs-all fashion.
@@ -274,7 +284,7 @@ declare class OrdinalEncoder {
  * // Result shape: [5, 3] with one-hot encoding
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelBinarizer.html | Scikit-learn LabelBinarizer}
+ * @see {@link https://deepbox.dev/docs/preprocess-encoders | Deepbox Encoders}
  */
 declare class LabelBinarizer {
     /** Indicates whether the binarizer has been fitted to data */
@@ -312,7 +322,7 @@ declare class LabelBinarizer {
      * @returns this - Returns self for method chaining
      * @throws {InvalidParameterError} If y is empty
      */
-    fit(y: Tensor): this;
+    fit(y: EncoderInput1D): this;
     /**
      * Transform labels to binary matrix.
      * Each label is converted to a binary vector with a single 1.
@@ -322,7 +332,7 @@ declare class LabelBinarizer {
      * @throws {NotFittedError} If binarizer is not fitted
      * @throws {InvalidParameterError} If y contains unknown labels
      */
-    transform(y: Tensor): Tensor | CSRMatrix;
+    transform(y: EncoderInput1D): Tensor | CSRMatrix;
     /**
      * Fit binarizer and transform labels in one step.
      * Convenience method equivalent to calling fit(y).transform(y).
@@ -330,7 +340,7 @@ declare class LabelBinarizer {
      * @param y - Target labels (1D tensor)
      * @returns Binary matrix (Tensor or CSRMatrix)
      */
-    fitTransform(y: Tensor): Tensor | CSRMatrix;
+    fitTransform(y: EncoderInput1D): Tensor | CSRMatrix;
     /**
      * Transform binary matrix back to labels.
      * Finds the column with maximum value for each row.
@@ -364,7 +374,7 @@ declare class LabelBinarizer {
  * // Each row can have multiple 1s
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MultiLabelBinarizer.html | Scikit-learn MultiLabelBinarizer}
+ * @see {@link https://deepbox.dev/docs/preprocess-encoders | Deepbox Encoders}
  */
 declare class MultiLabelBinarizer {
     /** Indicates whether the binarizer has been fitted to data */
@@ -449,7 +459,7 @@ declare class MultiLabelBinarizer {
  * const XScaled = scaler.transform(X);
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html | Scikit-learn StandardScaler}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class StandardScaler {
     private fitted;
@@ -480,7 +490,7 @@ declare class StandardScaler {
  *
  * **Formula**: X_scaled = (X - X.min) / (X.max - X.min) * (max - min) + min
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MinMaxScaler.html | Scikit-learn MinMaxScaler}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class MinMaxScaler {
     private fitted;
@@ -511,7 +521,7 @@ declare class MinMaxScaler {
  *
  * Scales to range [-1, 1]. Suitable for data that is already centered at zero.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MaxAbsScaler.html | Scikit-learn MaxAbsScaler}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class MaxAbsScaler {
     private fitted;
@@ -535,7 +545,7 @@ declare class MaxAbsScaler {
  *
  * Robust to outliers.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.RobustScaler.html | Scikit-learn RobustScaler}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class RobustScaler {
     private fitted;
@@ -572,7 +582,7 @@ declare class RobustScaler {
  *
  * Scales each sample (row) to have unit norm.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.Normalizer.html | Scikit-learn Normalizer}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class Normalizer {
     private norm;
@@ -596,7 +606,7 @@ declare class Normalizer {
  *
  * Maps to uniform or normal distribution.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.QuantileTransformer.html | Scikit-learn QuantileTransformer}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class QuantileTransformer {
     private fitted;
@@ -648,7 +658,7 @@ declare class QuantileTransformer {
  *
  * Supports Box-Cox and Yeo-Johnson transforms, with optional standardization.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.PowerTransformer.html | Scikit-learn PowerTransformer}
+ * @see {@link https://deepbox.dev/docs/preprocess-scalers | Deepbox Scalers}
  */
 declare class PowerTransformer {
     private fitted;
@@ -690,6 +700,13 @@ declare class PowerTransformer {
     fitTransform(X: Tensor): Tensor;
 }
 
+/**
+ * Represents a single train/test split with named index arrays.
+ */
+type SplitResult = {
+    readonly trainIndex: number[];
+    readonly testIndex: number[];
+};
 /**
  * Split arrays into random train and test subsets.
  *
@@ -712,7 +729,7 @@ declare class PowerTransformer {
  * const [XTrain, XTest, yTrain, yTest] = trainTestSplit(X, y, { testSize: 0.25 });
  * ```
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html | Scikit-learn train_test_split}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare function trainTestSplit(X: Tensor, y?: Tensor, options?: {
     testSize?: number;
@@ -726,7 +743,7 @@ declare function trainTestSplit(X: Tensor, y?: Tensor, options?: {
  *
  * Provides train/test indices to split data in train/test sets.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html | Scikit-learn KFold}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare class KFold {
     private nSplits;
@@ -737,7 +754,7 @@ declare class KFold {
         shuffle?: boolean;
         randomState?: number;
     });
-    split(X: Tensor): Array<[number[], number[]]>;
+    split(X: Tensor): SplitResult[];
     getNSplits(): number;
 }
 /**
@@ -745,7 +762,7 @@ declare class KFold {
  *
  * Provides train/test indices while preserving class distribution.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedKFold.html | Scikit-learn StratifiedKFold}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare class StratifiedKFold {
     private nSplits;
@@ -756,7 +773,7 @@ declare class StratifiedKFold {
         shuffle?: boolean;
         randomState?: number;
     });
-    split(X: Tensor, y: Tensor): Array<[number[], number[]]>;
+    split(X: Tensor, y: Tensor): SplitResult[];
     getNSplits(): number;
 }
 /**
@@ -764,34 +781,34 @@ declare class StratifiedKFold {
  *
  * Ensures same group is not in both train and test.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GroupKFold.html | Scikit-learn GroupKFold}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare class GroupKFold {
     private nSplits;
     constructor(options?: {
         nSplits?: number;
     });
-    split(X: Tensor, _y: Tensor | undefined, groups: Tensor): Array<[number[], number[]]>;
+    split(X: Tensor, _y: Tensor | undefined, groups: Tensor): SplitResult[];
     getNSplits(): number;
 }
 /**
  * Leave-One-Out cross-validator.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.LeaveOneOut.html | Scikit-learn LeaveOneOut}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare class LeaveOneOut {
-    split(X: Tensor): Array<[number[], number[]]>;
+    split(X: Tensor): SplitResult[];
     getNSplits(X: Tensor): number;
 }
 /**
  * Leave-P-Out cross-validator.
  *
- * @see {@link https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.LeavePOut.html | Scikit-learn LeavePOut}
+ * @see {@link https://deepbox.dev/docs/preprocess-splitting | Deepbox Data Splitting}
  */
 declare class LeavePOut {
     private p;
     constructor(p: number);
-    split(X: Tensor): Array<[number[], number[]]>;
+    split(X: Tensor): SplitResult[];
     getNSplits(X: Tensor): number;
 }
 
@@ -825,13 +842,14 @@ type index_QuantileTransformer = QuantileTransformer;
 declare const index_QuantileTransformer: typeof QuantileTransformer;
 type index_RobustScaler = RobustScaler;
 declare const index_RobustScaler: typeof RobustScaler;
+type index_SplitResult = SplitResult;
 type index_StandardScaler = StandardScaler;
 declare const index_StandardScaler: typeof StandardScaler;
 type index_StratifiedKFold = StratifiedKFold;
 declare const index_StratifiedKFold: typeof StratifiedKFold;
 declare const index_trainTestSplit: typeof trainTestSplit;
 declare namespace index {
-  export { index_GroupKFold as GroupKFold, index_KFold as KFold, index_LabelBinarizer as LabelBinarizer, index_LabelEncoder as LabelEncoder, index_LeaveOneOut as LeaveOneOut, index_LeavePOut as LeavePOut, index_MaxAbsScaler as MaxAbsScaler, index_MinMaxScaler as MinMaxScaler, index_MultiLabelBinarizer as MultiLabelBinarizer, index_Normalizer as Normalizer, index_OneHotEncoder as OneHotEncoder, index_OrdinalEncoder as OrdinalEncoder, index_PowerTransformer as PowerTransformer, index_QuantileTransformer as QuantileTransformer, index_RobustScaler as RobustScaler, index_StandardScaler as StandardScaler, index_StratifiedKFold as StratifiedKFold, index_trainTestSplit as trainTestSplit };
+  export { index_GroupKFold as GroupKFold, index_KFold as KFold, index_LabelBinarizer as LabelBinarizer, index_LabelEncoder as LabelEncoder, index_LeaveOneOut as LeaveOneOut, index_LeavePOut as LeavePOut, index_MaxAbsScaler as MaxAbsScaler, index_MinMaxScaler as MinMaxScaler, index_MultiLabelBinarizer as MultiLabelBinarizer, index_Normalizer as Normalizer, index_OneHotEncoder as OneHotEncoder, index_OrdinalEncoder as OrdinalEncoder, index_PowerTransformer as PowerTransformer, index_QuantileTransformer as QuantileTransformer, index_RobustScaler as RobustScaler, type index_SplitResult as SplitResult, index_StandardScaler as StandardScaler, index_StratifiedKFold as StratifiedKFold, index_trainTestSplit as trainTestSplit };
 }
 
-export { GroupKFold as G, KFold as K, LabelBinarizer as L, MultiLabelBinarizer as M, Normalizer as N, OneHotEncoder as O, PowerTransformer as P, QuantileTransformer as Q, RobustScaler as R, StandardScaler as S, LabelEncoder as a, OrdinalEncoder as b, MaxAbsScaler as c, MinMaxScaler as d, LeaveOneOut as e, LeavePOut as f, StratifiedKFold as g, index as i, trainTestSplit as t };
+export { GroupKFold as G, KFold as K, LabelBinarizer as L, MultiLabelBinarizer as M, Normalizer as N, OneHotEncoder as O, PowerTransformer as P, QuantileTransformer as Q, RobustScaler as R, StandardScaler as S, LabelEncoder as a, OrdinalEncoder as b, MaxAbsScaler as c, MinMaxScaler as d, LeaveOneOut as e, LeavePOut as f, type SplitResult as g, StratifiedKFold as h, index as i, trainTestSplit as t };